This work evaluates the possibility of replacing hydrogen sulfide (H2S) with thiosulfate anion (S2O32-) in sour corrosion fatigue studies. H2S increases the corrosion fatigue crack growth rate (FCGR) and can be present in carbon steel risers and flowlines used in off-shore oil production. Corrosion tests with gaseous H2S require special facilities with safety features, because H2S is a toxic and flammable gas. The possibility of replacing H2S with S2O32-, a non-toxic anion, for studying stress corrosion cracking of stainless and carbon steels in H2S solutions was first proposed by Tsujikawa et al. (Tsujikawa et al., Corrosion, 1993. 49(5): p. 409-419). In this dissertation, Tsujikawa work will be extended to sour corrosion fatigue of carbon steels.
H2S testing is often conducted in deareated condition to avoid oxygen reaction with sulfide that yields sulfur and to mimic oil production conditions. Nitrogen deareation was also adopted in S2O32- testing, and gas exiting the cell was forced through a sodium hydroxide trap. Measurements of the sulfide content of this trap were used to estimate the partial pressure of H2S in nitrogen, and Henry’s law was used to estimate the content of H2S in the solution in the cell. H2S was produced by a redox reaction of S2O32-, which required electrons from carbon steel corrosion. This reaction is spontaneous at the open circuit potential of steel. Therefore, H2S concentration was expected to be maximum at the steel surface, and this concentration was estimated by a mass balance analysis.
Carbon steel specimens exposed to S2O32- containing solutions developed a film on their surface, composed by iron sulfide and cementite. The film was not passivating and a good conductor of electrons. Hydrogen permeation experiments proved that this film controls the rate of hydrogen absorption of steels exposed to thiosulfate containing solutions. The absorption of hydrogen in S2O32- solutions was compared with the absorption of hydrogen in solutions saturated with different H2S partial pressures. The partial pressure was selected so that the concentration of H2S in the solution saturated with the gas would be the same as that reached in the surface of steel freely corroding in the thiosulfate solution. For solutions obtained by bubbling H2S, the rate of hydrogen absorption increased with the partial pressure of the gas, but the rate of hydrogen absorption reached a maximum at 10-3 M S2O32-, despite the surface concentration of H2S increased with the concentration of S2O32-. This effect was associated with the formation of thicker films, which inhibited the absorption of hydrogen.
FCGR were evaluated at constant stress intensity factor range. Crack length was monitored in-situ by the direct current potential drop (DCPD) method. FCGR increased with the partial pressure of H2S in nitrogen. FCGR was controlled not only by the amount of hydrogen present in the steel, but also by inhibiting contributions like crack closure and crack tip blunting. FCGR in dilute thiosulfate solutions was near that measured in a solution saturated with a partial pressure of H2S equal to 0.56 kPa, in accord with hydrogen permeation results.